Ab Initio Prediction of Borophene As An Extraordinary Anode Material Exhibiting Ultrafast Directional Sodium Diffusion For Sodium-Based Batteries

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Sci. Bull. (2016) 61(14):1138–1144 www.scibull.

com
DOI 10.1007/s11434-016-1118-7 www.springer.com/scp

Article Engineering Sciences

Ab initio prediction of borophene as an extraordinary anode


material exhibiting ultrafast directional sodium diffusion
for sodium-based batteries
Le Shi • Tianshou Zhao • Ao Xu • Jianbo Xu

Received: 9 May 2016 / Revised: 1 June 2016 / Accepted: 2 June 2016 / Published online: 18 June 2016
Ó Science China Press and Springer-Verlag Berlin Heidelberg 2016

Abstract Density functional theory calculations and 1 Introduction


ab initio molecular dynamics simulations are performed to
study the feasibility of using borophene, a newly synthe- Sodium-ion and sodium–oxygen batteries have recently
sized two-dimensional sheet of boron, as an anode material drawn considerable attention as a promising alternative to
for sodium-ion and sodium–oxygen batteries. The theo- lithium-based batteries due to the abundance and low price
retical capacity of borophene is found to be as high as of sodium element [1–6]. However, sodium is a highly
1,218 mAh g–1 (Na0.5B). More importantly, it is demon- reactive metal with low melting temperature, and also
strated that the sodium diffusion energy barrier along the shares the same dendrite problem with lithium, all of which
valley direction is as low as 0.0019 eV, which corresponds pose sever safety concern [7, 8]. Finding a suitable anode
to a diffusivity of more than a thousand times higher than material is one of the most urgent tasks before the com-
that of conventional anode materials such as Na2Ti3O7 and mercialization of sodium-based batteries. As the size of
Na3Sb. Hence, the use of borophene will revolutionize the sodium atom, 1.02 Å, is bigger than that of lithium,
rate capability of sodium-based batteries. Moreover, it is 0.76 Å, many of the anode materials that are good for
predicted that, during the sodiation process, the average lithium-based batteries are unsuitable for sodium-based
open-circuit voltage is 0.53 V, which can effectively sup- batteries. For example, graphite, a widely used anode
press the formation of dendrites while maximizing the material in lithium-ion batteries, showed extremely low
energy density. The metallic feature and structural integrity capacity when used as a sodium intercalation anode [1].
of borophene can be well preserved at different sodium Many possible materials, such as hard carbon, NiCo2O4
concentrations, demonstrating good electronic conductivity and Sn, have been tested for sodium-based batteries, but
and stable cyclability. these materials suffer from low intercalation utility, slow
kinetics and sever volume expansion [9–11]. Thus, finding
Keywords Borophene  Sodium anode  Directional a proper anode material with a large reversible capacity,
diffusion  Ultrafast diffusivity high sodium diffusion rate and good structural stability is
critically important for the development of sodium-based
batteries.
In recent years, two-dimensional (2D) materials draw
great attention as a new class of potential candidates for the
anode material of sodium-based batteries. The loose
Electronic supplementary material The online version of this packing between the 2D layers can accommodate the vol-
article (doi:10.1007/s11434-016-1118-7) contains supplementary
material, which is available to authorized users.
ume expansion caused by the insertion of sodium atoms
and maintain the structural integrity. Many 2D materials,
L. Shi  T. Zhao (&)  A. Xu  J. Xu such as defective/doped graphene [12, 13], transition metal
Department of Mechanical and Aerospace Engineering, The dichalcogenides (TMD) [14, 15], transition metal carbides
Hong Kong University of Science and Technology, Hong Kong,
China
(MXenes) [16, 17] and phosphorene [18, 19], have been
e-mail: [email protected] explored as potential candidates for the anode material of

123
Sci. Bull. (2016) 61(14):1138–1144 1139

sodium-based batteries using first-principle method, and Ea


D  e kB T ; ð2Þ
some of the predicted good performance have already been
proven in experiments [20, 21]. Recently, borophene, a 2D where Ea is the activation energy (diffusion barrier), kB is
sheet of boron, has been successfully synthesized by the Boltzmann constant and T is the environment
Mannix et al. [22]. The atomic structure of borophene was temperature.
resolved experimentally for the first time after many the- Ab initio molecular dynamics (AIMD) was used to
oretical predictions [23–25]. The 2D metallic feature and evaluate the structural stability of borophene and the dif-
low atomic weight of borophene make it an attractive fusion of sodium atom. The simulations were performed on
potential choice for the anode of sodium-based batteries. In a 5 9 3 borophene supercell at 300 K using NVT ensem-
this work, we comprehensively investigated the feasibility ble. The time step was set to be 1 fs and both simulations
of employing borophene as anode material for sodium- last 3 ps.
based batteries using density functional theory (DFT)
method. The calculation results show that borophene can
not only provide a superhigh capacity (1,218 mAh g–1), 3 Results and discussion
but also exhibit a directional ultrahigh sodium diffusivity,
which is estimated to be more than a thousand times 3.1 Structure of borophene
higher than that of conventional anode materials such as
Na2Ti3O7 [26] and Na3Sb [27] and one to seven magni- The top and side view of the optimized structure of bor-
tudes higher than other previously reported 2D materials ophene is shown in Fig. 1a, b. Different from the flat
[12–19]. The ultrahigh diffusivity will revolutionize the honeycomb structure of graphene [35], according to Man-
rate capability of sodium-based batteries. At different nix et al. [22], borophene belongs to the pmmn space group
sodium concentrations, the metallic feature and structural and shows a buckling structure. Here, we distinguish the
integrity of borophene can be well preserved, which boron atoms at different heights with ‘‘peak’’ and ‘‘valley’’,
ensure good electronic conductivity and stable cyclability. where ‘‘peak’’ denotes those atoms at higher positions in
The calculated average open-circuit voltage (OCV) is the z-axis considered, and ‘‘valley’’ denotes those atoms at
0.53 V, which is an appropriate value for sodium-based lower positions. The optimized lattice parameters are
batteries to maintain a high energy density while effec- a = 1.62 Å and b = 2.87 Å, which agree well with
tively suppressing the dendrite formation. All the calcu- experiment and previous calculation results [22, 36, 37].
lation results show that borophene is a prospective According to previous calculations [22, 36], the phonon
material for the anode of both sodium-ion and sodium– spectrum of free-standing borophene shows a small imag-
oxygen batteries. inary frequency near the C point, indicating its instability
against long-wavelength transversal waves, which can be
fixed by defects and may explain the observed stripe for-
2 Computational details mation in experiment. In this work, AIMD simulation was
performed to check the thermal stability of borophene at
All the calculations were performed using ABINIT soft- 300 K using a 5 9 3 supercell. The snapshot during the
ware package [28–30], with Perdew–Burke–Ernzerhof
(PBE) generalized gradient approximation (GGA) [31] and
projector augmented-wave (PAW) method [32]. The cut-
ting-off energy was set to be 20 Ha, and the k-point mesh
was set to be \0.01 Å-1. All the structures have been fully
optimized with a force tolerance of 0.01 eV Å-1. A 5 9 3
supercell was used to study the binding and diffusion of
sodium with a 20 Å vacuum spacing along the z-direction.
The binding energy was calculated as [33]
Eb ¼ EboropheneþNa  Eborophene  lNa ; ð1Þ
where Eborophene?Na is the total energy of the borophene
after bind with sodium, Eborophene is the total energy of
borophene and lNa is the chemical potential of metallic
sodium.
According to the Arrhenius equation [33, 34], the dif- Fig. 1 (Color online) a Optimized borophene structure from top view
fusion coefficient (D) of sodium can be estimated by and b side view, c the chosen adsorption sites on borophene

123
1140 Sci. Bull. (2016) 61(14):1138–1144

simulation process is shown in Fig. 2a. During the whole


simulation process (3 ps), no broken bonds were observed.
The movement of one boron atom was tracked as shown in
Fig. S1 (online). It can be found that the chosen boron atom
only vibrate around its initial position. The free energy of
the supercell during the simulation is shown in Fig. 2b, and
we can see the fluctuation is in a small range.

3.2 Adsorption of sodium onto borophene

To study the binding property of sodium on borophene,


four different binding sites with high symmetry (V1, V2,
P1, P2) have been investigated, as shown in Fig. 1c. The
calculated binding energies for sodium at different sites are
shown in Fig. 3a. All the binding energies have negative
values lower than -1.2 eV, indicating a strong binding
between sodium and borophene at all these four sites. The
negative binding energies indicate that the sodium atoms
tend to adsorb onto borophene uniformly instead of clus-
tering and forming a separate phase of sodium metal,
which can effectively suppress the formation of dendrites.
The valley sites show stronger binding than the peak sites,
while the energy differences between V1 and V2 or P1 and
P2 are quite small. V1 site shows the largest adsorption Fig. 3 (Color online) a Binding energies of sodium onto different
adsorption sites of borophene, b the charge transfer for a single
energy of -1.50 eV, which is larger than that for lithium sodium atom binding onto borophene at V1 site
atom (-1.12 eV) [37]. The charge transfer between sodium
and borophene when sodium adsorbed on V1 site is illus-
trated in Fig. 3b. The charge deficiency around sodium 3.3 Migration of sodium on the surface of borophene
atom and charge sufficiency around nearby boron atoms
indicate that the sodium atom loses its electrons to bor- Three diffusion routes were chosen to study the diffusion
ophene and exists in the form of cation. of sodium on borophene. As illustrated in Fig. 4a, these

Fig. 2 (Color online) a Snapshot from AIMD simulation of geometrical structure of borophene from top view and b side view, c evolution of
free energy for the borophene supercell during the simulation

123
Sci. Bull. (2016) 61(14):1138–1144 1141

three routes are labeled as ‘‘valley diffusion’’ (V1–V2–V1), AIMD simulation was also performed to evaluate the
‘‘peak diffusion 1’’ (V1–P2–V1) and ‘‘peak diffusion 2’’ thermal stability of borophene after adsorbed with sodium
(V1–P1–V1). In ‘‘valley diffusion’’, sodium diffuses along atom. Similar with that of pure borophene, during the
the ‘‘valleys’’, while in ‘‘peak diffusion’’, sodium diffuses whole simulation time (3 ps), no broken bonds were
across the ‘‘peaks’’. Nudged elastic band (NEB) calculation observed and the fluctuation of the free energy of the
was employed here to obtain the energy profiles. 9, 19 and system was in a small range as shown in Fig. S2a (online).
21 images were set for ‘‘valley diffusion’’, ‘‘peak diffusion More interestingly, when we track the movement of
1’’ and ‘‘peak diffusion 2’’ according to the diffusion adsorbed sodium atom, it can be found that the sodium
lengths. The obtained energy profiles are shown in Fig. 4b. atom moved along the x-direction (valley direction) as
Both ‘‘peak diffusion 1’’ and ‘‘peak diffusion 2’’ show a shown in Fig. S2b (online). The simulation results indicate
relative higher energy barrier of 0.24 and 0.25 eV, that sodium atoms could diffuse along the valley direction
respectively, while ‘‘valley diffusion’’ exhibits an ultralow freely at 300 K. Similar behavior can not be observed
energy barrier of 0.0019 eV, which agree well with the along the y-direction (peak direction), which agree well
above calculations on binding energies at different sites. with the anisotropic diffusion property calculated above.
This anisotropic diffusion property has also been predicted The high diffusion energy barrier along the peak direction
for lithium atom [37], which shows that the diffusion could also effectively confine the sodium atoms in the
energy barrier along the valley direction is a bit higher valleys and prevent them from clustering.
(0.0026 eV) than sodium.
According to Eq. (2), the diffusivity of sodium along the 3.4 Electrochemical properties of sodium adsorbed
valley direction at room temperature (T = 300 K) is about borophene
104 times higher than those along the peak directions,
which means the diffusion of sodium on borophene is To study the adsorption behavior at higher sodium con-
strongly anisotropic and the diffusion along the peak centrations, six different Na/B ratios (x = 0.0667, 0.0833,
directions are nearly impossible when compared with val- 0.125, 0.167, 0.25 and 0.5) have been considered by
ley diffusion. The diffusion energy barrier for sodium in employing 5 9 3, 4 9 3, 4 9 2 and 3 9 2 supercells. For
other investigated sodium anode materials such as each Na/B ratio, the sodium atoms were adsorbed onto both
Na2Ti3O7 and Na3Sb are 0.19 and 0.21 eV [27, 28], cor- sides of borophene evenly, as shown in Fig. 5a–f. When
responding to diffusivities about more than 10-3 times the concentration of sodium increases, as shown in Fig. 5g,
lower that of the ‘‘valley diffusion’’ in borophene. When the absolute value of binding energy will decrease
compared with other 2D materials, the energy barrier of accordingly due to the repulsion of adjacent sodium atoms,
‘‘valley diffusion’’ is still ultralow and the diffusivity is one while all the binding energies could keep negative lower
to seven magnitudes higher, as listed in Table S1 (online). than -0.5 eV. Thus, at high sodium concentrations, bor-
This ultrahigh diffusivity along the valley direction indi- ophene can still exhibit a strong binding toward sodium.
cates an excellent rate capability when using as anode The charge transfer between the binding sodium and bor-
material in sodium-based batteries. ophene is shown in Fig. S3 (online). In all the

Fig. 4 (Color online) a Three considered diffusion routes for sodium on the surface of borophene. b The energy profiles for the considered
diffusion routes

123
1142 Sci. Bull. (2016) 61(14):1138–1144

Fig. 5 (Color online) a–f The structures of NaxB with different sodium concentrations. g The binding energy of sodium on borophene with
different sodium concentrations

concentrations considered, the sodium atoms lose their both sides of sheet is shared, and the maximum concen-
electrons to borophene. When Na/B ratio reach 0.5, both tration will be reduced to half of the current estimation.
sides of borophene will be fully covered by the sodium, and The open-circuit voltage (OCV) can be estimated using
we regard this value to be the maximum capacity. The [38, 39]
calculated corresponding theoretical capacity of Na0.5B is ENax1 B  ENax2 B þ ðx2  x1 ÞENa
as high as 1,218 mAh g–1, much higher than those reported OCV  ; ð3Þ
ðx2  x1 Þe
in both experiments [9–11] and theoretical calculations
(listed in Table S1 online). The theoretical capacity of where ENax1 B ; ENax2 B and ENa are the energies of
borophene for sodium is a bit smaller than that for lithium Nax1 B, Nax2 B and metallic sodium, respectively. Figure 6a
[37] (Li0.75B, 1,860 mAh g–1), which is caused by the shows the OCV at various concentrations calculated using
larger atomic radius. In practice, the borophene sheet may Eq. (3). Similar with the absolute value of binding energy,
be not well separated. For stacked borophene, the Na on the OCV will decrease as the concentration of sodium

Fig. 6 (Color online) a The calculated voltage profile along the sodiation process. b The density of states (DOS) of borophene at different
sodium concentration. c The lattice parameters of borophene at different sodium concentration

123
Sci. Bull. (2016) 61(14):1138–1144 1143

atoms increases. All the OCVs are lower than 1.5 V, and References
when the Na/B ratio exceeds 0.125, the OCV will become
lower than 0.75 V. The calculated average OCV during the 1. Slater MD, Kim D, Lee E et al (2013) Sodium-ion batteries. Adv
Funct Mater 23:947–958
charge process is 0.53 V. This low theoretical OCV pro-
2. Pan H, Hu YS, Chen L (2013) Room-temperature stationary
vides the feasibility of using this material as the anode in sodium-ion batteries for large-scale electric energy storage.
sodium-ion and sodium–oxygen batteries, which could Energy Environ Sci 6:2338–2360
maintain a high energy density while suppressing the 3. Yabuuchi N, Kubota K, Dahbi M et al (2014) Research devel-
opment on sodium-ion batteries. Chem Rev 114:11636–11682
dendrite formation.
4. Han MH, Gonzalo E, Singh G et al (2015) A comprehensive
review of sodium layered oxides: powerful cathodes for Na-ion
3.5 Electronic properties and lattice deformation batteries. Energy Environ Sci 8:81–102
of borophene during the sodiation process 5. Xia C, Black R, Fernandes R et al (2015) The critical role of
phase-transfer catalysis in aprotic sodium oxygen batteries. Nat
Chem 7:496–501
The density of states (DOS) of bare borophene before and 6. Yadegari H, Banis MN, Xiao B et al (2015) Three-dimensional
after binding with sodium atoms at different concentrations nanostructured air electrode for sodium–oxygen batteries: a
are shown in Fig. 6b. All the considered systems are mechanism study toward the cyclability of the cell. Chem Mater
27:3040–3047
metallic, indicating a good electronic conductivity along
7. Chevrier VL, Ceder G (2011) Challenges for Na-ion negative
with the discharge–charge process. Figure 6c showed the electrodes. J Electrochem Soc 158:A1011–A1014
lattice parameters of borophene at different sodium con- 8. Mortazavi M, Deng J, Shenoy VB et al (2013) Elastic softening
centrations. Even at a concentration as high as Na0.5B, the of alloy negative electrodes for Na-ion batteries. J Power Sources
225:207–214
volume expansion is \2 %. The fluctuations of lattice
9. Ponrouch A, Goñi AR, Palacı́n MR (2013) High capacity hard
parameters along with the sodiation process are small, carbon anodes for sodium ion batteries in additive free elec-
indicating a good structural stability when using as anode trolyte. Electrochem Commun 27:85–88
material for sodium-based batteries. 10. Alcántara R, Jaraba M, Lavela P et al (2002) NiCo2O4 spinel:
First report on a transition metal oxide for the negative electrode
of sodium-ion batteries. Chem Mater 14:2847–2848
11. Liu Y, Zhang N, Jiao L et al (2015) Ultrasmall Sn nanoparticles
embedded in carbon as high-performance anode for sodium-ion
4 Conclusion batteries. Adv Funct Mater 25:214–220
12. Datta D, Li J, Shenoy VB (2014) Defective graphene as a high-
To summarize, we systematically investigated the feasi- capacity anode material for Na- and Ca-ion batteries. ACS Appl
bility of using the newly experimentally synthesized 2D Mater Interfaces 6:1788–1795
13. Ling C, Mizuno F (2014) Boron-doped graphene as a promising anode
metallic borophene as an anode material for sodium-based for Na-ion batteries. Phys Chem Chem Phys 16:10419–10424
batteries. The calculation results show that borophene 14. Mortazavi M, Wang C, Deng J et al (2014) Ab initio character-
exhibits a superhigh specific capacity (1,218 mAh g-1), ization of layered MoS2 as anode for sodium-ion batteries.
good structural stability and more importantly, an J Power Sources 268:279–286
15. Yang E, Ji H, Jung Y (2015) Two-dimensional transition metal
extraordinary high diffusivity along valley direction, which dichalcogenide monolayers as promising sodium ion battery
is expected to revolutionize the rate capability of sodium- anodes. J Phys Chem C 119:26374–26380
based batteries. The metallic feature of borophene can be 16. Yang E, Ji H, Kim J et al (2015) Exploring the possibilities of
well preserved along the whole sodiation process, ensuring two-dimensional transition metal carbides as anode materials for
sodium batteries. Phys Chem Chem Phys 17:5000–5005
good electronic conductivity. The calculated average OCV 17. Yu YX (2016) Prediction of mobility, enhanced storage capacity,
is 0.53 V, which is an appropriate value for sodium-ion and and volume change during sodiation on interlayer-expanded
sodium–oxygen batteries to suppress the dendrite forma- functionalized Ti3C2 MXene anode materials for sodium-ion
tion while maximizing the energy density. All the pre- batteries. J Phys Chem C 120:5288–5296
18. Kulish VV, Malyi OI, Persson C et al (2015) Phosphorene as an
sented results suggest that borophene is a promising anode material for Na-ion batteries: a first-principles study. Phys
material for the anode of sodium-based batteries. Chem Chem Phys 17:13921–13928
19. Hembram KP, Jung H, Yeo BC et al (2015) Unraveling the
Acknowledgment This work was fully supported by a Grant from atomistic sodiation mechanism of black phosphorus for sodium
the Research Grants Council of the Hong Kong Special Administra- ion batteries by first-principles calculations. J Phys Chem C
tive Region, China (16213414). 119:15041–15046
20. Xie X, Ao Z, Su D et al (2015) MoS2/Graphene composite anodes
Conflict of interest The authors declare that they have no conflict of with enhanced performance for sodium-ion batteries: the role of the
interest. two-dimensional heterointerface. Adv Funct Mater 25:1393–1403

123
1144 Sci. Bull. (2016) 61(14):1138–1144

21. Sun J, Lee HW, Pasta M et al (2015) A phosphorene–graphene 30. Gonze X (2005) A brief introduction to the ABINIT software
hybrid material as a high-capacity anode for sodium-ion batteries. package. Zeitschrift für Kristallographie Cryst Mater
Nat Nanotechnol 10:980–985 220:558–562
22. Mannix AJ, Zhou XF, Kiraly B et al (2015) Synthesis of bor- 31. Perdew JP, Burke K, Ernzerhof M (1996) Generalized gradient
ophenes: anisotropic, two-dimensional boron polymorphs. Sci- approximation made simple. Phys Rev Lett 77:3865
ence 350:1513–1516 32. Blöchl PE (1994) Projector augmented-wave method. Phys Rev B
23. Lau KC, Pandey R (2007) Stability and electronic properties of 50:17953
atomistically-engineered 2D boron sheets. J Phys Chem C 33. Jing Y, Zhou Z, Cabrera CR et al (2013) Metallic VS2 monolayer:
111:2906–2912 a promising 2D anode material for lithium ion batteries. J Phys
24. Penev ES, Bhowmick S, Sadrzadeh A et al (2012) Polymorphism Chem C 117:25409–25413
of two-dimensional boron. Nano Lett 12:2441–2445 34. Li W, Yang Y, Zhang G et al (2015) Ultrafast and directional
25. Piazza ZA, Hu HS, Li WL et al (2014) Planar hexagonal B36 as a diffusion of lithium in phosphorene for high-performance
potential basis for extended single-atom layer boron sheets. Nat lithium-ion battery. Nano Lett 15:1691–1697
Commun 5:3113 35. Neto AC, Guinea F, Peres NM et al (2009) The electronic
26. Pan H, Lu X, Yu X et al (2013) Sodium storage and transport properties of graphene. Rev Mod Phys 81:109
properties in layered Na2Ti3O7 for room-temperature sodium-ion 36. Peng B, Zhang H, Shao H et al (2016) The electronic, optical, and
batteries. Adv Energy Mater 3:1186–1194 thermodynamic properties of borophene from first-principles
27. Baggetto L, Ganesh P, Sun CN et al (2013) Intrinsic thermody- calculations. J Mater Chem C 4:3592–3598
namic and kinetic properties of Sb electrodes for Li-ion and Na- 37. Jiang HR, Lu Z, Wu MC et al (2016) Borophene: a promising
ion batteries: experiment and theory. J Mater Chem A anode material offering high specific capacity and high rate
1:7985–7994 capability for lithium-ion batteries. Nano Energy 23:97–104
28. Gonze X, Beuken JM, Caracas R et al (2002) First-principles 38. Ling C, Mizuno F (2012) Capture lithium in aMnO2: insights
computation of material properties: the ABINIT software project. from first principles. Chem Mater 24:3943–3951
Comput Mater Sci 25:478–492 39. Aydinol MK, Kohan AF, Ceder G et al (1997) Ab initio study of
29. Gonze X, Amadon B, Anglade PM et al (2009) ABINIT: first- lithium intercalation in metal oxides and metal dichalcogenides.
principles approach to material and nanosystem properties. Phys Rev B 56:1354
Comput Phys Commun 180:2582–2615

123

You might also like